Summary Our research focuses on the regulation of gene expression, in particular the mechanisms controlling cellular and viral mRNA expression. A critical step in the mRNA metabolism is the transport of the mRNA from the nucleus to the cytoplasm. Analysis of retroviral systems, pioneered by research on HIV-1, have shed light into some important aspects of nuclear mRNA export and have provided critical insights into mechanisms governing cellular mRNA and protein transport. We are utilizing retroviral systems to identify and study mechanisms of mRNA metabolism using a combination of biochemistry, functional genomics, and proteomics. The dissection of the mechanisms of posttranscriptional control and nucleocytoplasmic trafficking of macromolecules are relevant to understand processes involved in cellular gene expression as well as virus expression. We identified the mRNA export requirement of the simian type D retroviral transcript which is mediated by the cis-acting RNA export element (CTE) and its binding partner, the cellular protein NXF1. We further found that the cellular NXF1 protein acts as the key nuclear receptor for cellular mRNAs, and that this function is conserved in metazoa. We identified that the expression and mobility of the murine LTR-retroelements (Intracisternal A particle retroelements) depends on the presence of the cis-acting RNA transport element RTE, which acts like the CTE. This finding reveals that, despite a complex evolutionary history, retroelements and retroviruses share the dependency on posttranscriptional regulation. We identified the RNA binding motif 15 (RBM15) protein as the cellular factor that binds and exports RTE-containing mRNAs via the NXF1 export pathway. RBM15, a novel mRNA export factor, belongs to the SPEN family of proteins and is conserved among metazoa. We further identified that another SPEN protein, OTT3 acts as RNA export co-factor like RBM15. On the other hand, SHARP, another SPEN family member does not have export function. Biochemical and subcellular localization studies showed that OTT3 and RBM15 also interact with each other in vivo, further supporting a shared function. Genetic knock-down of RBM15 in mouse is embryonic lethal, indicating that OTT3 cannot compensate for the RBM15 loss which supports the notion that these proteins, in addition to sharing similar activities, likely have distinct biological roles. Why vertebrates had evolved and maintain variants of the otherwise similar factors? Since RBM15 and OTT3 are factors of the NXF1 pathway, which controls the export of general mRNA, we speculated that they may be part of a developmental or/and tissue-specific switch that controls mRNA export rates or/and specificity. Importantly, we found that RBM15 and OTT3 act as molecular link to the NXF1 export pathway. Thus, our studies on retroviral RNA export led to identification important factors of the mRNA export route. We also studied regulation of expression of cytokine genes. The use of cytokine DNAs (IL-12 and IL-15) as molecular vaccine adjuvants was found to improve the quantity and alter the quality of the immune responses. To optimally use these cytokines, we studied their regulation and found that IL-15/IL-15Ra as well as the IL-12 cytokine family use similar posttranscriptional and posttranslational regulation strategies. As a result, the formation and secretion of the subunits and heterodimers are highly regulated steps. Using this knowledge, we have generated optimized expression vectors, which allow their efficient use in animal models, and their use as molecular adjuvant in vaccine and in cancer immunotherapy is promising and could be important for future translational applications. IL-15 expression is controlled at several posttranscriptional and posttranslational steps such as mRNA stability, translation, intracellular trafficking and secretion. The combination of posttranscriptional and posttranslational modification led to the generation of efficient expression plasmids producing several hundred fold higher levels of bioactive IL-15. There are two known isoforms of IL-15 containing either a long signal peptide (LSP) or a short signal peptide (SSP), and are produced by alternatively spliced transcripts. We have studied the function of these 2 variants of IL-15. We found that similar to LSP IL-15, the SSP IL-15 is stabilized and secreted efficiently upon coexpression of IL-15Ralpha. Coinjection of SSP IL-15- and IL-15Ralpha-expressing plasmids into mice resulted in increased plasma levels of bioactive heterodimeric IL-15 and mobilization and expansion of NK and T cells. Therefore, SSP IL-15 is secreted and bioactive when produced as a heterodimer with IL-15Ralpha in the same cell. The apparent half-life of this heterodimer is lower compared with LSP IL-15/IL-15Ralpha, due to different intracellular processing. Coexpression of both LSP IL-15 and SSP IL-15 in the presence of IL-15Ralpha results in lower levels of bioactive IL-15, indicating that LSP and SSP IL-15 compete for the binding to IL-15Ralpha when expressed in the same cell. Because the SSP IL-15 interaction to IL-15Ralpha leads to a complex with lower apparent stability, SSP IL-15 functions as competitive inhibitor of LSP IL-15. The data suggest that usage of alternative splicing is an additional level of control of IL-15 activity. Expression of both SSP and LSP forms of IL-15 appears to be conserved in many mammals, suggesting that SSP may be important for expressing a form of IL-15 with lower magnitude or duration of biological effects. We also study the family of IL-12 and related cytokines, which consist of 2 polypeptide chains. Our studies revealed an important aspect of the posttranslational control of these cytokines. Using the expression of individual subunits alone or in combinations, we found that one of the subunits was relatively unstable and limiting;expression of the other subunit in the same cell increased the stability and secretion of the unstable partner and the levels of the secreted heterodimer. This allowed use to generate optimized vectors to produce such cytokines for in vivo application. As an application in a cancer model, injection of optimized IL-27 DNA together with IL-2 successfully abolished neuroblastoma metastasis. Future studies will be directed to replacing IL-2 with other cytokines like IL-15/IL-15Ra in such cancer models.